Discover carrier-grade reliability with specifically hardened NFVI Blueprint solutions from HPE.
I think we can all agree on one thing: Downtime stinks. Your customers get angry when your services don't work as expected, your partners wonder how reliable you are in holding up your end of the bargain, and your employees resent having to work overtime to resolve problems.
But for certain verticals, downtime is more than a nuisanceโit simply isn't an option. Carrier-grade network function virtualization (NFV) customers need proven solutions that meet the highest levels of certification, because high availability is paramount. Often, lives depend on services being active and ready to go, especially after disasters and other events that might render ordinary hardware useless.
With features such as reliability, high availability, and performance being mission critical in a telecommunications environment, the carrier-grade NFV solution you choose should be comprised of performance-optimized hardware and software that conforms to carrier-grade requirements.
The good news? Our telecom cloud solutions meet this criteria.
Enhanced hardware for all
Our telecom cloud solutions are based on proven mainstream HPE hardware. When telecoms or other specific verticals require us to certify our servers to meet certain industry requirements, we ensure that any necessary improvements are built into the mainstream compute product. That means all of our customers get the benefit of the hardening and enhancements we make for any one specific industry, at prices that reflect the economies of scale available to organizations choosing commodity hardware.
In the case of the telecommunications industry, many of our carrier partners require that our hardware meet the NEBS Level 3 guidelines.
This means our equipment can, among other things:
o Resist high (or low) temperature and humidity, vibrations, particles, or fumes in the air
o Minimize the risk of a fire starting or spreading
o Help with space planning and installation
o Use electricity more efficiently than standard equipment
o Handle electrical disturbances and radio frequency interference without skipping a beat
All of those aspects make our equipment truly carrier-grade.
Fast, efficient compute
Being carrier-grade means choosing best-of-breed components, even at the compute layer. Our servers don't use just any Intel CPUs; rather, we specifically choose Intel CPUs designed to have a long life, engineered with telco service providers and other demanding applications in mind.
We also choose CPUs that support the Data Plane Development Kit (DPDK), a program that significantly improves packet processing performance and throughput. This allows more system resources for high-throughput data plane applications. Additionally, the CPUs we select support single root input/output virtualization (SR/IOV), a specification that lets a single PCI Express resource appear as multiple discrete resources. This way, it's simpler to virtualize access to that device across multiple virtualization platforms, ensuring enhanced performance.
Express data plane
The data plane is a critical component of any cloud-oriented stack. We test our network adapters against the highest bandwidth and throughput standards, and verify their support for key virtualization and management standards. Ultimately, we've found that our solution performs at a level that would generally be associated with custom, expensive, third-party-specific purpose boxes and appliances. Thesesolutions have unmatched performance in virtualization and virtual switching, with market-leading latency and failover characteristics.
Our carrier-grade NFV telecom cloud solution has specially tested and curated components that allow extreme packet acceleration in the data plane, with DPDK-enabled NIC cards and DPDK-enabled chipsets containing support for SRIOV and PCI pass-through (PCI-PT). The firmware and drivers for all of our network adapters are tested against these high standards, as well. For other solutions, this sometimes represents a stumbling block, since this type of testing rigor and the ensuing feedback cycle require a very specialized set of skills and expertise.โWe have the experience and dedication to ensure our network layer components are not just capable of performing at 10 Gig wire speeds, but also at 25, and up to 100 Gig.
The hypervisor and software side
We pay special attention to the software stack we put on these telco solutions to make sure the operating system and the virtual infrastructure manager (VIM) layer provides carrier-grade reliability as well.
A key part of this layer is the HPE Red Hat OpenStack Platform, which is HPE's commercially supported distribution of Red Hat OpenStack virtual infrastructure management software. We designed this platform to build and manage large pools of compute, storage, and networking resources for cloud deployments. The HPE Red Hat OpenStack distribution is optimized to run on HPE open hardware infrastructure equipped with Red Hat Enterprise Linux (RHEL), the industry's most popular Linux operating system.
We also configured our blueprints to deploy the software in a way that specifically incorporates best design practices, with no single points of failure and a carrier-grade VIM that truly operates at a "five 9s" level of reliability. With this equipment in your office, you'll always be a step ahead of yourcompetition.
Explore our solutions
You can depend on our carrier-grade NFV infrastructure platform to perform at the highest levels. To learn more, visit our website to explore the latest NFVI Blueprint solutions.
Claus_Pedersen
Business leader for HPEโs Telco Segment and Network Function Virtualization Solutions in HPEโs Hybrid IT Business group. Responsible for go-to market activities, product development and strategic partnerships enabling HPEโs infrastructure business in the telecommunication segment.
